Bacteria evolve around antibiotic trade-offs, threatening a promising resistance strategy
Researchers discovered that bacteria can quickly overcome collateral sensitivity—a strategy where resistance to one antibiotic makes bacteria weaker to another. The finding suggests this approach to extending antibiotic lifespans may be less durable than hoped, forcing drug developers to rethink how to combat the growing crisis of resistant infections.
Originaltitel: Mechanistic Origins and Evolutionary Erosion of Collateral Sensitivity in a b-lactamase
<p>As antibiotic discovery stalls, exploiting collateral sensitivity, where resistance to one drug increases sensitivity to another, offers a promising route to extend the lifespan of existing drugs. However, the molecular origins and robustness of such trade-offs at the level of single resistance determinants remain poorly understood. Here, we examined a previously evolved trajectory of the b-lactamase OXA-48 to Q4 (A33V/F72L/T212A/S213A) in Escherichia coli. Compared to OXA-48, Q4 conferred 40-fold increase in ceftazidime resistance, but a 27-fold lower piperacillin resistance. This trade-off was caused by the introduction of F72L. We challenged the stability of this collateral sensitivity network by subjecting Q4 to directed evolution followed by co-selection from both ceftazidime and piperacillin. The emerging substitution V120G alleviated the piperacillin trade-off while maintaining elevated resistance to ceftazidime in genetic backgrounds harboring F72L. Structural and computational analyses revealed that evolution introduced substantial conformational changes in the X-loop, likely leading to less productive piperacillin binding poses. V120G counteracted the effect of F72L by decreasing the X-loop's conformational freedom, thereby partially restoring piperacillin resistance. Finally, we show that other substitutions at position 120 can exert similar mitigating effects. Taken together, our results provide a mechanistic understanding of how adaptive solutions both generate and erode collateral sensitivity, knowledge crucial for predicting the long-term stability of these networks. (c) 2026 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecom-mons.org/licenses/by/4.0/).</p>